AI-Enabled Infrastructure Could Better Resist And Recover From Mass Outages
A US federal government project will receive multiple millions of dollars in the coming years to investigate ways to use AI to better predict, respond to, and ultimately avoid disruptions in the US power infrastructure. Ultimately, researchers seek to make an “autonomous” grid feasible that uses AI technologies to improve the efficiency and reduce vulnerabilities in the grid. IEEE Spectrum reports:
A new project, called GRIP, for Grid Resilience and Intelligence Project, was awarded up to $6 million over three years on September 12 by the U.S. Department of Energy. GRIP is the first project to use artificial intelligence (AI) to help power grids deal with disturbances, says Sila Kiliccote, GRIP’s principal investigator and director of the Grid Integration, Systems and Mobility lab at the SLAC National Accelerator Laboratory in Menlo Park, Calif.
GRIP will develop algorithms to learn how power grids work by analyzing smart meter data, utility-scale SCADA (supervisory control and data acquisition) data, electric vehicle charging data, and even satellite and street-view imagery.
The aim with GRIP is to address three different kinds of problems. “First we need to anticipate and get in front of grid events,” Kiliccote says. “Next we’d like to minimize the effects of grid events when they do happen. Finally, after the event ends, we’d want to bring systems back as quickly as possible.”
GRIP’s first year is devoted to anticipating grid problems. Predictive analytics will help identify places where the electric grid is vulnerable to disruption so it can be reinforced, Kiliccote says.
The second year will aim to help grids absorb disruptions. For instance, a grid can be divided into virtual “islands,” or micro-grids, that can be isolated to prevent a power disruption from spreading and taking down the entire grid.
The third year will focus on helping grids recover from events. “Ultimately, we’d like to see a grid that can run on its own, an autonomous grid like an autonomous car,” Kiliccote says. “However, unlike autonomous cars, an autonomous grid will need to be able to handle additional components added to it while it is still running.”